Add optional second base argument to int DSL function (#1244)

* Add optional second base argument to `int` DSL function

* artifacts from `make dev`

* unit-test cases

* typofix

docs/src/manpage.md

@@ -2388,7 +2388,15 @@ MILLER(1)                                                            MILLER(1)
         (class=system #args=0) Returns the hostname as a string.
 
    1mint0m
-        (class=conversion #args=1) Convert int/float/bool/string to int.
+        (class=conversion #args=1,2) Convert int/float/bool/string to int. If the second argument is omitted and the first argument is a string, base is inferred from the first argument's prefix. If the second argument is provided and the first argument is a string, the second argument is used as the base. If the second argument is provided and the first argument is not a string, the second argument is ignored.
+       Examples:
+       int("345") gives decimal 345 (base-10/decimal input is inferred)
+       int("0xff") gives decimal 255 (base-16/hexadecimal input is inferred)
+       int("0377") gives decimal 255 (base-8/octal input is inferred)
+       int("0b11010011") gives decimal 211 which is hexadecimal 0xd3 (base-2/binary input is inferred)
+       int("0377", 10) gives decimal 377
+       int(345, 16) gives decimal 345
+       int(string(345), 16) gives decimal 837
 
    1minvqnorm0m
         (class=math #args=1) Inverse of normal cumulative distribution function. Note that invqorm(urand()) is normally distributed.
@@ -3338,5 +3346,5 @@ MILLER(1)                                                            MILLER(1)
 
 
 
-                                  2023-03-12                         MILLER(1)
+                                  2023-03-24                         MILLER(1)
 </pre>

docs/src/manpage.txt

@@ -2367,7 +2367,15 @@ MILLER(1)                                                            MILLER(1)
         (class=system #args=0) Returns the hostname as a string.
 
    1mint0m
-        (class=conversion #args=1) Convert int/float/bool/string to int.
+        (class=conversion #args=1,2) Convert int/float/bool/string to int. If the second argument is omitted and the first argument is a string, base is inferred from the first argument's prefix. If the second argument is provided and the first argument is a string, the second argument is used as the base. If the second argument is provided and the first argument is not a string, the second argument is ignored.
+       Examples:
+       int("345") gives decimal 345 (base-10/decimal input is inferred)
+       int("0xff") gives decimal 255 (base-16/hexadecimal input is inferred)
+       int("0377") gives decimal 255 (base-8/octal input is inferred)
+       int("0b11010011") gives decimal 211 which is hexadecimal 0xd3 (base-2/binary input is inferred)
+       int("0377", 10) gives decimal 377
+       int(345, 16) gives decimal 345
+       int(string(345), 16) gives decimal 837
 
    1minvqnorm0m
         (class=math #args=1) Inverse of normal cumulative distribution function. Note that invqorm(urand()) is normally distributed.
@@ -3317,4 +3325,4 @@ MILLER(1)                                                            MILLER(1)
 
 
 
-                                  2023-03-12                         MILLER(1)
+                                  2023-03-24                         MILLER(1)

docs/src/reference-dsl-builtin-functions.md

@@ -547,7 +547,15 @@ hexfmt  (class=conversion #args=1) Convert int to hex string, e.g. 255 to "0xff"
 
 ### int
 <pre class="pre-non-highlight-non-pair">
-int  (class=conversion #args=1) Convert int/float/bool/string to int.
+int  (class=conversion #args=1,2) Convert int/float/bool/string to int. If the second argument is omitted and the first argument is a string, base is inferred from the first argument's prefix. If the second argument is provided and the first argument is a string, the second argument is used as the base. If the second argument is provided and the first argument is not a string, the second argument is ignored.
+Examples:
+int("345") gives decimal 345 (base-10/decimal input is inferred)
+int("0xff") gives decimal 255 (base-16/hexadecimal input is inferred)
+int("0377") gives decimal 255 (base-8/octal input is inferred)
+int("0b11010011") gives decimal 211 which is hexadecimal 0xd3 (base-2/binary input is inferred)
+int("0377", 10) gives decimal 377
+int(345, 16) gives decimal 345
+int(string(345), 16) gives decimal 837
 </pre>
 
 

internal/pkg/bifs/types.go

@@ -55,6 +55,53 @@ func BIF_int(input1 *mlrval.Mlrval) *mlrval.Mlrval {
 	return to_int_dispositions[input1.Type()](input1)
 }
 
+// ----------------------------------------------------------------
+func string_to_int_with_base(input1, input2 *mlrval.Mlrval) *mlrval.Mlrval {
+	i, ok := lib.TryIntFromStringWithBase(input1.AcquireStringValue(), input2.AcquireIntValue())
+	if ok {
+		return mlrval.FromInt(i)
+	} else {
+		return mlrval.ERROR
+	}
+}
+
+func int_to_int_with_base(input1, input2 *mlrval.Mlrval) *mlrval.Mlrval {
+	return mlrval.FromInt(int64(input1.AcquireIntValue()))
+}
+
+func float_to_int_with_base(input1, input2 *mlrval.Mlrval) *mlrval.Mlrval {
+	return mlrval.FromInt(int64(input1.AcquireFloatValue()))
+}
+
+func bool_to_int_with_base(input1, input2 *mlrval.Mlrval) *mlrval.Mlrval {
+	if input1.AcquireBoolValue() == true {
+		return mlrval.FromInt(1)
+	} else {
+		return mlrval.FromInt(0)
+	}
+}
+
+var to_int_with_base_dispositions = [mlrval.MT_DIM]BinaryFunc{
+	/*INT    */ int_to_int_with_base,
+	/*FLOAT  */ float_to_int_with_base,
+	/*BOOL   */ bool_to_int_with_base,
+	/*VOID   */ _void,
+	/*STRING */ string_to_int_with_base,
+	/*ARRAY  */ _erro,
+	/*MAP    */ _erro,
+	/*FUNC   */ _erro,
+	/*ERROR  */ _erro,
+	/*NULL   */ _null,
+	/*ABSENT */ _absn,
+}
+
+func BIF_int_with_base(input1, input2 *mlrval.Mlrval) *mlrval.Mlrval {
+	if !input2.IsInt() {
+		return mlrval.ERROR
+	}
+	return to_int_with_base_dispositions[input1.Type()](input1, input2)
+}
+
 // ----------------------------------------------------------------
 func string_to_float(input1 *mlrval.Mlrval) *mlrval.Mlrval {
 	f, ok := lib.TryFloatFromString(input1.AcquireStringValue())

internal/pkg/dsl/cst/builtin_function_manager.go

@@ -1529,10 +1529,25 @@ Note that NaN has the property that NaN != NaN, so you need 'is_nan(x)' rather t
 		},
 
 		{
-			name:      "int",
-			class:     FUNC_CLASS_CONVERSION,
-			help:      "Convert int/float/bool/string to int.",
-			unaryFunc: bifs.BIF_int,
+			name:  "int",
+			class: FUNC_CLASS_CONVERSION,
+			help: `Convert int/float/bool/string to int.
+If the second argument is omitted and the first argument is a string, base is inferred from the first argument's prefix.
+If the second argument is provided and the first argument is a string, the second argument is used as the base.
+If the second argument is provided and the first argument is not a string, the second argument is ignored.`,
+
+			unaryFunc:          bifs.BIF_int,
+			binaryFunc:         bifs.BIF_int_with_base,
+			hasMultipleArities: true,
+			examples: []string{
+				`int("345") gives decimal 345 (base-10/decimal input is inferred)`,
+				`int("0xff") gives decimal 255 (base-16/hexadecimal input is inferred)`,
+				`int("0377") gives decimal 255 (base-8/octal input is inferred)`,
+				`int("0b11010011") gives decimal 211 which is hexadecimal 0xd3 (base-2/binary input is inferred)`,
+				`int("0377", 10) gives decimal 377`,
+				`int(345, 16) gives decimal 345`,
+				`int(string(345), 16) gives decimal 837`,
+			},
 		},
 
 		{

internal/pkg/lib/util.go

@@ -122,6 +122,27 @@ func TryIntFromString(input string) (int64, bool) {
 	return 0, false
 }
 
+// TryIntFromStringWithBase allows the user to choose the base that's used,
+// rather than inferring from 0x prefix, etc as TryIntFromString does.
+func TryIntFromStringWithBase(input string, base int64) (int64, bool) {
+	// Go's strconv parses "1_2" as 12; not OK for Miller syntax. (Also not valid JSON.)
+	for i := 0; i < len(input); i++ {
+		if input[i] == '_' {
+			return 0, false
+		}
+	}
+
+	i64, ierr := strconv.ParseInt(input, int(base), 64)
+	if ierr == nil {
+		return i64, true
+	}
+	u64, uerr := strconv.ParseUint(input, int(base), 64)
+	if uerr == nil {
+		return int64(u64), true
+	}
+	return 0, false
+}
+
 func TryFloatFromString(input string) (float64, bool) {
 	// Go's strconv parses "1_2.3_4" as 12.34; not OK for Miller syntax. (Also not valid JSON.)
 	for i := 0; i < len(input); i++ {

man/manpage.txt

@@ -2367,7 +2367,15 @@ MILLER(1)                                                            MILLER(1)
         (class=system #args=0) Returns the hostname as a string.
 
    1mint0m
-        (class=conversion #args=1) Convert int/float/bool/string to int.
+        (class=conversion #args=1,2) Convert int/float/bool/string to int. If the second argument is omitted and the first argument is a string, base is inferred from the first argument's prefix. If the second argument is provided and the first argument is a string, the second argument is used as the base. If the second argument is provided and the first argument is not a string, the second argument is ignored.
+       Examples:
+       int("345") gives decimal 345 (base-10/decimal input is inferred)
+       int("0xff") gives decimal 255 (base-16/hexadecimal input is inferred)
+       int("0377") gives decimal 255 (base-8/octal input is inferred)
+       int("0b11010011") gives decimal 211 which is hexadecimal 0xd3 (base-2/binary input is inferred)
+       int("0377", 10) gives decimal 377
+       int(345, 16) gives decimal 345
+       int(string(345), 16) gives decimal 837
 
    1minvqnorm0m
         (class=math #args=1) Inverse of normal cumulative distribution function. Note that invqorm(urand()) is normally distributed.
@@ -3317,4 +3325,4 @@ MILLER(1)                                                            MILLER(1)
 
 
 
-                                  2023-03-12                         MILLER(1)
+                                  2023-03-24                         MILLER(1)

man/mlr.1

@@ -2,12 +2,12 @@
 .\"     Title: mlr
 .\"    Author: [see the "AUTHOR" section]
 .\" Generator: ./mkman.rb
-.\"      Date: 2023-03-12
+.\"      Date: 2023-03-24
 .\"    Manual: \ \&
 .\"    Source: \ \&
 .\"  Language: English
 .\"
-.TH "MILLER" "1" "2023-03-12" "\ \&" "\ \&"
+.TH "MILLER" "1" "2023-03-24" "\ \&" "\ \&"
 .\" -----------------------------------------------------------------
 .\" * Portability definitions
 .\" ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
@@ -3362,7 +3362,15 @@ gsub("prefix4529:suffix8567", "(....ix)([0-9]+)", "[\e1 : \e2]") gives "[prefix
 .RS 0
 .\}
 .nf
- (class=conversion #args=1) Convert int/float/bool/string to int.
+ (class=conversion #args=1,2) Convert int/float/bool/string to int. If the second argument is omitted and the first argument is a string, base is inferred from the first argument's prefix. If the second argument is provided and the first argument is a string, the second argument is used as the base. If the second argument is provided and the first argument is not a string, the second argument is ignored.
+Examples:
+int("345") gives decimal 345 (base-10/decimal input is inferred)
+int("0xff") gives decimal 255 (base-16/hexadecimal input is inferred)
+int("0377") gives decimal 255 (base-8/octal input is inferred)
+int("0b11010011") gives decimal 211 which is hexadecimal 0xd3 (base-2/binary input is inferred)
+int("0377", 10) gives decimal 377
+int(345, 16) gives decimal 345
+int(string(345), 16) gives decimal 837
 .fi
 .if n \{\
 .RE

test/cases/dsl-int-function/0001/cmd

@@ -0,0 +1 @@
+mlr --nidx --from ${CASEDIR}/input -S put -f ${CASEDIR}/mlr

test/cases/dsl-int-function/0001/expout

@@ -0,0 +1,4 @@
+345 345
+0xff 255
+0377 255
+0b11010011 211

test/cases/dsl-int-function/0001/input

@@ -0,0 +1,4 @@
+345
+0xff
+0377
+0b11010011

test/cases/dsl-int-function/0001/mlr

@@ -0,0 +1 @@
+$2 = int($1) + 0

test/cases/dsl-int-function/0002/cmd

@@ -0,0 +1 @@
+mlr --nidx --from ${CASEDIR}/input -S put -f ${CASEDIR}/mlr

test/cases/dsl-int-function/0002/expout

@@ -0,0 +1,4 @@
+345 345
+ff (error)
+0377 377
+11010011 11010011

test/cases/dsl-int-function/0002/input

@@ -0,0 +1,4 @@
+345
+ff
+0377
+11010011

test/cases/dsl-int-function/0002/mlr

@@ -0,0 +1 @@
+$2 = int($1, 10) + 0

test/cases/dsl-int-function/0003/cmd

@@ -0,0 +1 @@
+mlr --nidx --from ${CASEDIR}/input -S put -f ${CASEDIR}/mlr

test/cases/dsl-int-function/0003/expout

@@ -0,0 +1,4 @@
+345 837
+ff 255
+0377 887
+11010011 285278225

test/cases/dsl-int-function/0003/input

@@ -0,0 +1,4 @@
+345
+ff
+0377
+11010011

test/cases/dsl-int-function/0003/mlr

@@ -0,0 +1 @@
+$2 = int($1, 16) + 0

test/cases/dsl-int-function/0004/cmd

@@ -0,0 +1 @@
+mlr --nidx --from ${CASEDIR}/input -S put -f ${CASEDIR}/mlr

test/cases/dsl-int-function/0004/expout

@@ -0,0 +1,4 @@
+345 229
+ff (error)
+0377 255
+11010011 2363401

test/cases/dsl-int-function/0004/input

@@ -0,0 +1,4 @@
+345
+ff
+0377
+11010011

test/cases/dsl-int-function/0004/mlr

@@ -0,0 +1 @@
+$2 = int($1, 8) + 0

test/cases/dsl-int-function/0005/cmd

@@ -0,0 +1 @@
+mlr --nidx --from ${CASEDIR}/input -S put -f ${CASEDIR}/mlr

test/cases/dsl-int-function/0005/expout

@@ -0,0 +1,4 @@
+345 (error)
+ff (error)
+0377 (error)
+11010011 211

test/cases/dsl-int-function/0005/input

@@ -0,0 +1,4 @@
+345
+ff
+0377
+11010011

test/cases/dsl-int-function/0005/mlr

@@ -0,0 +1 @@
+$2 = int($1, 2) + 0

test/cases/dsl-int-function/0006/cmd

@@ -0,0 +1 @@
+mlr --nidx --from ${CASEDIR}/input put -f ${CASEDIR}/mlr

test/cases/dsl-int-function/0006/expout

@@ -0,0 +1,4 @@
+345 345
+ff (error)
+0377 255
+11010011 11010011

test/cases/dsl-int-function/0006/input

@@ -0,0 +1,4 @@
+345
+ff
+0377
+11010011

test/cases/dsl-int-function/0006/mlr

@@ -0,0 +1 @@
+$2 = int($1, 8) + 0